X-ray ct apparatus, medical image processing apparatus, and medical image processing method

Abstract

An X-ray CT apparatus according to an embodiment includes a photographing system configured to photograph a subject, and processing circuitry. The processing circuitry generates X-ray CT image data by performing reconstruction processing on projection data output from the photographing system, calculates an image feature quantity indicating a feature of a relation between one pixel on the X-ray CT image data and one or a plurality of pixels other than the former pixel, specifies a contribution region indicating a region on the X-ray CT image data delineating a feature having a high influence degree for a calculation result of the image feature quantity, and causes a display apparatus to display the X-ray CT image data representing the contribution region.

What is claimed is: 
     
         1 . An X-ray CT apparatus comprising:
 a photographing system configured to photograph a subject; and   processing circuitry configured to
 generate X-ray CT image data by performing reconstruction processing on projection data output from the photographing system, 
 calculate an image feature quantity indicating a feature of a relation between one pixel on the X-ray CT image data and one or a plurality of pixels other than the former pixel, 
 specify a contribution region indicating a region on the X-ray CT image data delineating the feature having a high influence degree for a calculation result of the image feature quantity, and 
 cause a display apparatus to display the X-ray CT image data representing the contribution region. 
   
     
     
         2 . The X-ray CT apparatus according to  claim 1 , wherein
 the processing circuitry
 calculates an element feature quantity as a feature quantity corresponding to each of features on the X-ray CT image data for each of the features, and calculates a first influence degree indicating a degree of influence of the feature on a calculation result of the image feature quantity based on the image feature quantity and the element feature quantity, and 
 specifies the contribution region based on the first influence degree. 
   
     
     
         3 . The X-ray CT apparatus according to  claim 2 , wherein the processing circuitry specifies, as the contribution region, a region corresponding to the feature having a highest first influence degree on the X-ray CT image data. 
     
     
         4 . The X-ray CT apparatus according to  claim 2 , wherein the processing circuitry specifies, as the contribution region, a region corresponding to the feature having the first influence degree exceeding a threshold on the X-ray CT image data. 
     
     
         5 . The X-ray CT apparatus according to  claim 2 , wherein
 the processing circuitry
 estimates a disease condition of the subject based on the image feature quantity and a pre-trained model caused to have a function of outputting an estimation result of a disease condition in response to an input of the image feature quantity by machine learning, and 
 causes the display apparatus to display the X-ray CT image data representing the contribution region as grounds for estimation of the estimation result. 
   
     
     
         6 . The X-ray CT apparatus according to  claim 5 , wherein
 the processing circuitry
 calculates a plurality of types of the image feature quantities, 
 calculates the first influence degree for each of a plurality of types of the image feature quantities, calculates, for each pixel on the X-ray CT image data, a second influence degree indicating a degree of influence of the pixel on each of a plurality of types of the image feature quantities based on the first influence degree and a weight indicating how much weight is given to a specific type of the image feature quantity when the pre-trained model outputs the estimation result, and specifies the contribution region based on the second influence degree. 
   
     
     
         7 . The X-ray CT apparatus according to  claim 1 , wherein the processing circuitry specifies, based on the specified contribution region and an image database storing therein a plurality of pieces of the X-ray CT image data representing the contribution region, the X-ray CT image data with similar distribution of contribution regions on the X-ray CT image data in the image database, and causes the specified X-ray CT image data to be displayed as a reference image. 
     
     
         8 . A medical image processing apparatus comprising:
 processing circuitry configured to
 acquire a medical image, 
 calculate an image feature quantity indicating a feature of a relation between one pixel on the medical image and one or a plurality of pixels other than the former pixel, 
 specify a contribution region indicating a region on the medical image delineating the feature having a high influence degree for a calculation result of the image feature quantity, and 
 cause a display apparatus to display the medical image representing the contribution region. 
   
     
     
         9 . The medical image processing apparatus according to  claim 8 , wherein
 the processing circuitry
 calculates an element feature quantity as a feature quantity corresponding to each of features on the medical image for each of the features, and calculates a first influence degree indicating a degree of influence of the feature on a calculation result of the image feature quantity based on the image feature quantity and the element feature quantity, and 
   specifies the contribution region based on the first influence degree.   
     
     
         10 . The medical image processing apparatus according to  claim 9 , wherein the processing circuitry specifies, as the contribution region, a region corresponding to the feature having a highest first influence degree on the medical image. 
     
     
         11 . The medical image processing apparatus according to  claim 9 , wherein the processing circuitry specifies, as the contribution region, a region corresponding to the feature having the first influence degree exceeding a threshold on the medical image. 
     
     
         12 . The medical image processing apparatus according to  claim 9 , wherein
 the processing circuitry
 estimates a disease condition of the subject based on the image feature quantity and a pre-trained model caused to have a function of outputting an estimation result of a disease condition in response to an input of the image feature quantity by machine learning, and 
 causes the display apparatus to display the medical image representing the contribution region as grounds for estimation of the estimation result. 
   
     
     
         13 . The medical image processing apparatus according to  claim 12 , wherein
 the processing circuitry
 calculates a plurality of types of the image feature quantities, 
 calculates the first influence degree for each of a plurality of types of the image feature quantities, 
 calculates, for each pixel on the medical image, a second influence degree indicating a degree of influence of the pixel on each of a plurality of types of the image feature quantities based on the first influence degree and a weight indicating how much weight is given to a specific type of the image feature quantity when the pre-trained model outputs the estimation result, and 
 specifies the contribution region based on the second influence degree. 
   
     
     
         14 . The medical image processing apparatus according to  claim 8 , wherein the processing circuitry specifies, based on the specified contribution region and an image database storing therein a plurality of the medical images representing the contribution region, the medical image with similar distribution of contribution regions on the medical image in the image database, and causes the specified medical image to be displayed as a reference image. 
     
     
         15 . A medical image processing method performed by a medical image processing apparatus, the medical image processing method comprising:
 acquiring a medical image;   calculating an image feature quantity indicating a feature of a relation between one pixel on the medical image and one or a plurality of pixels other than the former pixel,   specifying a contribution region indicating a region on the medical image delineating a feature having a high influence degree for a calculation result of the image feature quantity, and   causing a display apparatus to display the medical image representing the contribution region.